INDEX

Explanations

The

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

((&

-0.29

 acts

-0.28

Act

-0.28

Act

-0.27

itis

-0.27

then

-0.27

act

-0.26

something

-0.25

experimental

-0.25

assume

-0.25

POSITIVE LOGITS

åı²

0.26

-meta

0.26

 Malta

0.26

WithValue

0.26

Guild

0.24

ä»£çłģ

0.24

 Slayer

0.24

 Specific

0.24

(jQuery

0.24

Specific

0.24

Activations Density 0.040%

The

No Known Activations